The 2004 federal election may be remembered as much for re-introducing minority government to Parliament as it will for its political advertising. While few, if any, elections are won or lost on one variable, the Liberals use of negative advertising in this one may be an important piece of the electoral puzzle. Understanding how they used advertising, as well as its relationship within the overall campaign might go some way to understanding the Liberal victory.

Observers usually find several defining moments in a campaign and the federal election this year was no differ- ent. Among the usual list of things that are seen to change the course of an election are debates, advertising and the leaders’ tours. Both leaders’ tours and debates are heavily interpreted by the media. Our perception of them is shaped by media, pundits and spin doctors. Advertising is one of the few areas in which parties can control their messages but ”” as we shall see ”” often with little preci- sion. As one of the campaign’s most visible and constant factors, advertising is seen as the perennial barometer of an election. Showing its high and low pressure points and predicting ways in which the electoral climate may change, political advertising may be an important indica- tor of who won and why.

As a form of communication, advertising works on a number of dif- ferent levels. Its effects can be seen not only in terms of paid media, that is the size of the advertising budget, but perhaps more so in terms of earned media, the amount of atten- tion the ads garner in the media.

When we think about the effective- ness of political ads, we need to think not only about whether it directly changed voters’ minds but also about the ripples it generated in the media pond. Not only do ads work directly and indirectly, their effects on us can be seen in several ways. Good politi- cal ads provide both intellectual as well as emotional food for thought. But great ads are those which are burned into our psyche and resonate at an emotional level through power- ful visuals. These mind bombs tap into existing thoughts often by mak- ing arguments that force the reader to supply a missing premise or piece of evidence. In doing so, we become agents of our own persuasion.

Political campaigns ”” both here and abroad ”” are rife with examples of evocative visuals being used in election advertising. These ads are, almost exclusively, negative ones which might suggest that while we continue to reject negative ads, we do remember them. The begin- ning of political ads on television was 1952, though the real turning point was 1964 with the (in)famous  ”œDaisy.” So incendiary was this ad, that it was broadcast only once because of its stark claims. It depict- ed a little girl picking petals from a flower as an ominous voice counted down from ten. It is quickly appar- ent that the voice is counting down for a rocket launch. As the voice gets to ”œone” the camera zooms into the girls eye which transforms into a nuclear mushroom cloud. Lyndon Johnson intones ”œThese are the stakes. To make a world in which all of God’s children can live or to go into the darkness. We must either love each other or we must die.” The narrator says ”œVote for President Johnson on November 3rd. The stakes are too high for you to stay home.” The ad makes a powerful claim about Johnson’s opponent, Barry Goldwater, without ever refer- ring to him by name. More recently, Democratic contender Michael Dukakis was ridiculed in a 1988 ad that had him driving around in a tank. His Republican opponent, George H.W. Bush, used the footage to attack the claim that Dukakis was a strong supporter of national defence. That same election gave us the ”œWilly Horton” ad whose visual of a revolving turnstile powerfully reinforced the verbal argument that was being made, namely that Dukakis was soft on crime. In all of these cases, the ad did not supply any new information but was effec- tive at reinforcing existing perceptions ”” a fact that bears remember- ing this election.

In response to these ads, Canadians have argued smugly that our poli- tics was different. We would never use such blatantly personal negative ads. Our recent history has seemed to bear this out until, of course, the ill-fated ”œIs this a prime minister?” ad that the Progressive Conserva- tives ran in 1993. Showing Jean Chretien’s facial paralysis the ads stated ”œI personally would be very embarrassed if he were to become prime minister of Canada.” While this was seen as an unequivocal failure, it marked the beginning of advertising as signal flare: a technique to get media and therefore public attention.

Notwithstanding this one, our use of negative political ads has been more focused on issues than personal attributes. The textbook example of this was the Liberal ”œborder ad” that ran in 1988 after the leaders’ debates. In a few sentences the tenor of the federal election changed dramatically. The ad showed two men sitting on opposite sides of a table. The Canada and US flags on the table identify them as negotiators. As the deal is about to be finalized, the American says, ”œ…there’s one line I’d like to change.” The earnest young Canadian negotiator replies, ”œWhich line is that?” A pencil eraser begins to rub out the 49th parallel and the American says, ”œThis line here. It’s just getting in the way.” The ad worked because it picked up on polling data that many Canadians were concerned that free trade would result in a loss of Canadian sovereignty.

Demonstrating the power of responding to negative ads, the Conservatives responded with a car- pet-bombing strategy that featured a small number of ads shown in very high rotation. The ads visually and verbally had drawn the border line back in, while the narrator closed with, ”œthis is where we draw the line.” In doing so, the valence of the origi- nal Liberal ad was muted and the effects were softened. The Liberals actually withdrew their border ad ear- lier than scheduled.

One of the intriguing questions raised in the 2004 election is why the Conservatives relied largely on posi- tive ads and did not respond in kind to the rather powerful negative claims made in Liberal advertising. With only a couple of exceptions, the Conservative ads were largely positive.

Negative ads are rarely used by themselves but rather are an adjunct to a broader campaign strategy. In the 1988 free trade election, the ads were only a part of the wider cam- paign to sow seeds of doubt about the merits of free trade or the moti- vations of its detractors. When then finance minister Michael Wilson announced that the Americans would rip up the Auto Pact if free trade was not ratified, advertising reinforced that message. We need to see advertising as the tip of the cam- paign iceberg. While it might be the most visible element of the cam- paign, its presence tells more about the large mass that is hidden than what is visible.

In this election, predictably all the major parties used advertising to support their campaign’s key mes- sages. In the case of the Liberals, their negative ads mimicked their cam- paign. Their strategy of asking Canadians to ”œthink again” before voting Conservative was not merely confined to their advertising but rather was evident in campaign speeches, photo-ops and other forms of communication. The Conservatives played the single note of government profligacy and waste ”” embodied in their slogan of ”œdemand better.” While their ads largely supported that campaign theme, it simply did not have the visceral power of the Liberal’s. An appeal to our pocket books was not as strong as an appeal to our hearts.

The Conservatives did broadcast negative ads. Two of them, ”œCookie Jar” and ”œCarousel,” had simple story lines. Neither had the requisite emo- tional connection of the Liberal ads and both relied on the same argument that the Liberal government was wasteful. ”œCookie Jar” showed a cuffed hand ”” presumably representative of the Liberals ”” taking cookies out of a jar and another hand (the Conservatives?) putting them back in. Relying on carnival music that was dissonant with the visuals, ”œCarousel” showed stacks of money being thrown out, with the tag line, ”œCan you really afford another four years of Liberal waste?” Its humour was timid, its critiques repetitive and, most importantly, there was no clear attribution of responsibility to the govern- ment. It’s not a surprise that it was not as memorable as Liberal ads. Polling done by Leger Marketing earlier in the cam- paign supported this. They found that 74 percent of Canadians remember seeing one of the Liberal ads, 60 pecent a Conservative ad and 42 percent had seen one from the NDP.

The Liberal ad that generated the most heat in the election was first aired on June 9, at the halfway point of the campaign. At this stage in the campaign, Conservative support was marginally higher than the Liberals but was trending upward according to polls done by CPAC-SES and Ipsos-Reid. ”œHarper and the Conservatives” was produced by the same ad company that brought us Molson’s ”œThe Rant.” Like ”œThe Rant,” ”œHarper and the Conservatives” made its point using a series of powerful, iconic images. Beginning with the provocative image of tanks rolling through a desert, mov- ing on to a gun pointing at the camera through to a woman rocking on the floor of a hospital, the ad juxtaposed menacing images with an ominous soundtrack. It ended with a female voice-over that said ”œStephen Harper says that when he’s through with Canada, we won’t recognize it. You know what? He’s right.”

A second Liberal ad, ”œConservative Economics,” superimposed the face of Stephen Harper over images of Brian Mulroney and Mike Harris. Using the same ominous music, the ad used an old technique of transference to make a visual argument (Harper was like Mulroney and Harris) without any kind evidence. In fact, the only connection verbally was ”œNow Stephen Harper wants his turn.” It was strikingly similar to the NDP ad that had Harper’s face between Ernie Eves and Mike Harris with the tag line ”œConservative Cuts and Privatization.” Both worked in the same manner as the Daisy ad did years ago ”” by making connections without any evidence to support them.

One of the problems of attempting to infer the effects of an ad in an elec- tion campaign is that it is difficult to separate the ad’s effects from the media’s reporting of the ad and campaign photo-ops that sup- port the arguments of the ad. If ”œHarper and the Conservatives” was successful, then it was so in part because it was an element of the larger Liberal campaign that highlighted the negative attrib- utes of voting Conservative.

The argument made in that ad ”” that Harper and the Conservatives were extreme and had a secret agenda ”” may have also been reinforced by the Conservative candidates themselves. Claims were made by candidates such as Randy White who said a Conservative government would scrap same- sex marriages and ban abor- tions. Cheryl Gallant was quoted as saying that abortion was no different than the Iraqi beheading of American con- tractor Nicholas Berg and that gays and lesbians ought to be excluded from hate laws.

Harper did not do enough to distance his party from these comments and also may have reinforced them himself. When, on June 18, the Conservative leader issued a press release saying ”œPaul Martin Supports Child Pornography?” the comparisons to the press release of Ontario Premier Ernie Eves that likened Dalton McGuinty to a ”œEvil Reptilian Kitten Eater” were obvious. Three days later he said that his government would revise the legis- lation that mandated French on Air Canada flights causing a further brush- fire that needed to be stamped out. In all of these cases, the Conservatives were providing superb earned media for Liberal claims found in their ads.

The fact that the Conservatives did not answer the Liberal arguments is surprising given how critical they were. They may not have responded for a number of reasons. Perhaps they were not anticipating the claims in them, though this seems implausible as they were supported by other cam- paign material. A more likely scenario is that they believed that, like other negative ads, they would backfire. Why they did not had to do with the way in which images were framed. Jack Bensimon, the creator of ”œHarper and the Conservatives” summarizes it well when he says ”œfear of the alterna- tive trumped anger at the status quo.”

Like the ”œfacial paralysis” ad of Jean Chrétien in 1993, the Conservatives may have thought that it was sufficiently personal to not war- rant a response.

Could another explanation be that the Conservatives thought that they had won the election and therefore needn’t have responded to them? The Liberals use of negative ads suggest that they knew that it was a tight election. The last weekend cross-Canada sprint by Paul Martin reaffirms his claim that he was indeed in a ”œfight for his life.” This is in sharp contrast to Stephen Harper, who spent the late part of the campaign in the safe confines of Western Canada ”” hardly evidence that they believed the election was in trouble. Paul Wells, writing in Maclean’s, says that with 48 hours to go, Stephen Harper spoke about the election in the past tense, as if it were a done deal.

Parties have not yet reported their official expenditures to Elections Canada, but media buying numbers sup- port the claim that in this election all three national parties spent more on advertising than in previous elections. Taking advantage of the Elections Act, which excludes expenses made before the writ, both the Liberals and the Conservatives began their advertising campaign before the election with elec- tion-style spots featuring both their respective leaders. Preliminary data from Neilsen are available on some election advertising. The data from Toronto show that the Conservatives engaged in more advertising than the other parties prior to the election and that during the elec- tion they outspent the other parties. This is likely a result of the Conservatives spending more in prime time than the Liberals, whose ads ran in higher fre- quency in the greater Toronto area. In Toronto, the NDP had the second-high- est gross rating points of all parties, which supports the oft-made claim that the NDP was also heavily involved in the advertising game in this election.

While the NDP generally was on the sidelines of the main advertising battle, this election was pivotal for them for at least a couple of reasons. First is the breakthrough in the amount of money it spent on advertising ”” likely an unprecedented amount for that party. Second is the decision by the NDP to run an ad on ethnic chan- nels featuring Jack Layton and Olivia Chow speaking in Mandarin. This was the first time a party had taken such a step. While parties routinely translate paper campaign material into different languages, and take out ads in ethnic newspapers, the step to produce a tele- vision ad in a non-official language was entirely new. This trend was also pioneered in 2000 when Republican presiden- tial candidates aired televi- sion ads in Spanish to appeal to the Hispanic pop- ulation. Both of these might also suggest that the NDP is moving more toward the American-style campaigning of the other two parties and a brokerage style of politics, which sees the party adopting centrist policies to appeal to diverse, regionally concentrated ethnic communities.

To ask if election ads were effective is to somehow see a simple correla- tion between an advertisement and popular support. The way in which ads work is usually more complex than that. Good negative ads do not per- suade as much as they are able to rein- force existing opinion and translate that into sowing seeds of doubt about one’s opponent. While attempting to infer election results from advertising is a bit like reading tea leaves, it appears as if the Liberal ads did just that ”” they raised sufficient enough questions about Stephen Harper and the Conservative’s agenda. What this elec- tion reminds us is that in a closely fought contest, advertising can play an important role in shaping the outcome.


Jonathan Rose would like to acknowledge the assistance of Simon Kiss and Hugh Mellon, both of whom provided helpful comments.

You are welcome to republish this Policy Options article online or in print periodicals, under a Creative Commons/No Derivatives licence.

Creative Commons License